add logging

app.py

@@ -65,6 +65,7 @@ def inference(lm_model_name: str, cls_model_name: str, prompt: str, fp16: bool =
     else:
         autocast = torch.cuda.amp.autocast
     with autocast(fp16):
+        print(f"Generating for prompt: {prompt}")
         sequences, tokens = generator.sample_sequences(
             num_samples=1,
             input_prompt=prompt,
@@ -74,6 +75,7 @@ def inference(lm_model_name: str, cls_model_name: str, prompt: str, fp16: bool =
             entropy=3.2,
             **kwargs
         )
+        print(f"Output for prompt: {sequences}")
     return sequences[0]